Diffusion-Weighted Images Superresolution Using High-Order SVD

Figure 1

Principle eigenvectors for the tensor model applied to a synthetic phantom. (a) The original dataset, the phantom datasets reconstructed using (b) B-splines, (c) the nonlocal method, (d) the proposed HOSVD, and (e) the proposed HOSVD-M. (f) The noisy phantom (SNR = 30) and the phantom dataset reconstructed using (g) B-splines, (h) the nonlocal method, (i) the proposed HOSVD, and (j) the proposed HOSVD-M.

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

(j)